Multiview: a software package for multiview pattern recognition methods

SUMMARY

Multiview datasets are the norm in bioinformatics, often under the label multi-omics. Multiview data are gathered from several experiments, measurements or feature sets available for the same subjects. Recent studies in pattern recognition have shown the advantage of using multiview methods of clustering and dimensionality reduction; however, none of these methods are readily available to the extent of our knowledge. Multiview extensions of four well-known pattern recognition methods are proposed here. Three multiview dimensionality reduction methods: multiview t-distributed stochastic neighbour embedding, multiview multidimensional scaling and multiview minimum curvilinearity embedding, as well as a multiview spectral clustering method. Often they produce better results than their single-view counterparts, tested here on four multiview datasets.

AVAILABILITY AND IMPLEMENTATION

R package at the B2SLab site: http://b2slab.upc.edu/software-and-tutorials/ and Python package: https://pypi.python.org/pypi/multiview.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Effects of Repeated Sublethal External Exposure to Deep Water Horizon Oil on the Avian Metabolome

We assessed adverse effects of external sublethal exposure of Deepwater Horizon, Mississippi Canyon 252 oil on plasma and liver metabolome profiles of the double-crested cormorant (Phalacrocorax auritus), a large (1.5 to 3.0 kg) diving waterbird common in the Gulf of Mexico. Metabolomics analysis of avian plasma showed significant negative effects on avian metabolic profiles, in some cases after only two external exposures (26 g cumulative) to oil. We observed significant (p < 0.05) changes in intermediate metabolites of energy metabolism and fatty acid and amino acid metabolic pathways in cormorants after repeated exposure to oil. Exposure to oil increased several metabolites (glycine, betaine, serine and methionine) that are essential to the one-carbon metabolism pathway. Lipid metabolism was affected, causing an increase in production of ketone bodies, suggesting lipids were used as an alternative energy source for energy production in oil exposed birds. In addition, metabolites associated with hepatic bile acid metabolism were affected by oil exposure which was correlated with changes observed in bile acids in exposed birds. These changes at the most basic level of phenotypic expression caused by sublethal exposure to oil can have effects that would be detrimental to reproduction, migration, and survival in avian species.

Metabolic Reprogramming by 3-Iodothyronamine (T1AM): A New Perspective to Reverse Obesity through Co-Regulation of Sirtuin 4 and 6 Expression

Obesity is a complex disease associated with environmental and genetic factors. 3-Iodothyronamine (T1AM) has revealed great potential as an effective weight loss drug. We used metabolomics and associated transcriptional gene and protein expression analysis to investigate the tissue specific metabolic reprogramming effects of subchronic T1AM treatment at two pharmacological daily doses (10 and 25 mg/kg) on targeted metabolic pathways. Multi-analytical results indicated that T1AM at 25 mg/kg can act as a novel master regulator of both glucose and lipid metabolism in mice through sirtuin-mediated pathways. In liver, we observed an increased gene and protein expression of Sirt6 (a master gene regulator of glucose) and Gck (glucose kinase) and a decreased expression of Sirt4 (a negative regulator of fatty acids oxidation (FAO)), whereas in white adipose tissue only Sirt6 was increased. Metabolomics analysis supported physiological changes at both doses with most increases in FAO, glycolysis indicators and the mitochondrial substrate, at the highest dose of T1AM. Together our results suggest that T1AM acts through sirtuin-mediated pathways to metabolically reprogram fatty acid and glucose metabolism possibly through small molecules signaling. Our novel mechanistic findings indicate that T1AM has a great potential as a drug for the treatment of obesity and possibly diabetes.

Clear differences in cerebrospinal fluid proteome between women with chronic widespread pain and healthy women - a multivariate explorative cross-sectional study

Introduction

Frequent chronic local pain can develop into chronic widespread pain (CWP). The spread of pain is correlated with pain intensity, anxiety, and depression, conditions that ultimately lead to a poor quality of life. Knowledge is incomplete about CWP’s etiology, although it has been suggested that both central hyperexcitability and/or a combination with peripheral factors may be involved. Cerebrospinal fluid (CSF) could act as a mirror for the central nervous system as proteins are signal substances that activate the formation of algesics and control nociceptive processes. To this end, this study investigates the CSF protein expression in women with CWP and in female healthy controls.

Materials and methods

This study included 12 female patients with CWP diagnosed according to the American College of Rheumatology criteria with 13 healthy age- and sex-matched pain-free subjects. All subjects went through a clinical examination and answered a health questionnaire that registered sociodemographic and anthropometric data, pain characteristics, psychological status, and quality of life rating. CSF was collected by lumbar puncture from each subject. Two-dimensional gel electrophoresis in combination with mass spectrometry was used to analyze the CSF proteome. This study identifies proteins that significantly discriminate between the two groups using multivariate data analysis (MVDA) (i.e., orthogonal partial least squares discriminant analysis [OPLS-DA]).

Results

There were no clinically significant levels of psychological distress and catastrophization presented in subjects with CWP. MVDA revealed a highly significant OPLS-DA model where 48 proteins from CSF explained 91% (R²) of the variation and with a prediction of 90% (Q²). The highest discriminating proteins were metabolic, transport, stress, and inflammatory.

Conclusion

The highest discriminating proteins (11 proteins), according to the literature, are involved in apoptotic regulations, anti-inflammatory and anti-oxidative processes, the immune system, and endogenous repair. The results of this explorative study may indicate the presence of neuro-inflammation in the central nervous system of CWP patients. Future studies should be larger and control for confounders and determine which alterations are unspecific/general and which are specific changes.

Metabolic profiling reveals reprogramming of lipid metabolic pathways in treatment of polycystic ovary syndrome with 3-iodothyronamine

Complex diseases such as polycystic ovary syndrome (PCOS) are associated with intricate pathophysiological, hormonal, and metabolic feedbacks that make their early diagnosis challenging, thus increasing the prevalence risks for obesity, cardiovascular, and fatty liver diseases. To explore the crosstalk between endocrine and lipid metabolic pathways, we administered 3-iodothyronamine (T1AM), a natural analog of thyroid hormone, in a mouse model of PCOS and analyzed plasma and tissue extracts using multidisciplinary omics and biochemical approaches. T1AM administration induces a profound tissue-specific antilipogenic effect in liver and muscle by lowering gene expression of key regulators of lipid metabolism, PTP1B and PLIN2, significantly increasing metabolites (glucogenic, amino acids, carnitine, and citrate) levels, while enhancing protection against oxidative stress. In contrast, T1AM has an opposing effect on the regulation of estrogenic pathways in the ovary by upregulating STAR, CYP11A1, and CYP17A1. Biochemical measurements provide further evidence of significant reduction in liver cholesterol and triglycerides in post-T1AM treatment. Our results shed light onto tissue-specific metabolic vs. hormonal pathway interactions, thus illuminating the intricacies within the pathophysiology of PCOS This study opens up new avenues to design drugs for targeted therapeutics to improve quality of life in complex metabolic diseases.

Nonlinear Dimensionality Reduction by Minimum Curvilinearity for Unsupervised Discovery of Patterns in Multidimensional Proteomic Data

Dimensionality reduction is largely and successfully employed for the visualization and discrimination of patterns, hidden in multidimensional proteomics datasets. Principal component analysis (PCA), which is the preferred approach for linear dimensionality reduction, may present serious limitations, in particular when samples are nonlinearly related, as often occurs in several two-dimensional electrophoresis (2-DE) datasets. An aggravating factor is that PCA robustness is impaired when the number of samples is small in comparison to the number of proteomic features, and this is the case in high-dimensional proteomic datasets, including 2-DE ones. Here, we describe the use of a nonlinear unsupervised learning machine for dimensionality reduction called minimum curvilinear embedding (MCE) that was successfully applied to different biological samples datasets. In particular, we provide an example where we directly compare MCE performance with that of PCA in disclosing neuropathic pain patterns, hidden in a multidimensional proteomic dataset.

Multivariate proteomic analysis of the cerebrospinal fluid of patients with peripheral neuropathic pain and healthy controls - a hypothesis-generating pilot study

Pain medicine lacks objective biomarkers to guide diagnosis and treatment. Combining two-dimensional gel proteomics with multivariate data analysis by projection, we exploratively analyzed the cerebrospinal fluid of eleven patients with severe peripheral neuropathic pain due to trauma and/or surgery refractory to conventional treatment and eleven healthy controls. Using orthogonal partial least squares discriminant analysis, we identified a panel of 36 proteins highly discriminating between the two groups. Due to a possible confounding effect of age, a new model with age as outcome variable was computed for patients (n=11), and four out of 36 protein spots were excluded due to a probable influence of age. Of the 32 remaining proteins, the following seven had the highest discriminatory power between the two groups: an isoform of angiotensinogen (upregulated in patients), two isoforms of alpha-1-antitrypsin (downregulated in patients), three isoforms of haptoglobin (upregulated in patients), and one isoform of pigment epithelium-derived factor (downregulated in patients). It has recently been hypothesized that the renin–angiotensin system may play a role in the pathophysiology of neuropathic pain, and a clinical trial of an angiotensin II receptor antagonist was recently published. It is noteworthy that when searching for neuropathic pain biomarkers with a purely explorative methodology, it was indeed a renin–angiotensin system protein that had the highest discriminatory power between patients and controls in the present study. The results from this hypothesis-generating pilot study have to be confirmed in larger, hypothesis-driven studies with age-matched controls, but the present study illustrates the fruitfulness of combining proteomics with multivariate data analysis in hypothesis-generating pain biomarker studies in humans.

Nonlinear dimension reduction and clustering by Minimum Curvilinearity unfold neuropathic pain and tissue embryological classes

MOTIVATION

Nonlinear small datasets, which are characterized by low numbers of samples and very high numbers of measures, occur frequently in computational biology, and pose problems in their investigation. Unsupervised hybrid-two-phase (H2P) procedures-specifically dimension reduction (DR), coupled with clustering-provide valuable assistance, not only for unsupervised data classification, but also for visualization of the patterns hidden in high-dimensional feature space.

METHODS

'Minimum Curvilinearity' (MC) is a principle that-for small datasets-suggests the approximation of curvilinear sample distances in the feature space by pair-wise distances over their minimum spanning tree (MST), and thus avoids the introduction of any tuning parameter. MC is used to design two novel forms of nonlinear machine learning (NML): Minimum Curvilinear embedding (MCE) for DR, and Minimum Curvilinear affinity propagation (MCAP) for clustering.

RESULTS

Compared with several other unsupervised and supervised algorithms, MCE and MCAP, whether individually or combined in H2P, overcome the limits of classical approaches. High performance was attained in the visualization and classification of: (i) pain patients (proteomic measurements) in peripheral neuropathy; (ii) human organ tissues (genomic transcription factor measurements) on the basis of their embryological origin.

CONCLUSION

MC provides a valuable framework to estimate nonlinear distances in small datasets. Its extension to large datasets is prefigured for novel NMLs. Classification of neuropathic pain by proteomic profiles offers new insights for future molecular and systems biology characterization of pain. Improvements in tissue embryological classification refine results obtained in an earlier study, and suggest a possible reinterpretation of skin attribution as mesodermal.

AVAILABILITY

https://sites.google.com/site/carlovittoriocannistraci/home.

On-chip technologies for multidimensional separations

We review microfluidic devices designed for multidimensional sample analysis, with a primer on relevant theory, an emphasis on protein analysis, and an eye towards future improvements and challenges to the field. Image shows results of an on-chip IEF-CE separation of a protein mixture; unpublished surface plot data from A. E. Herr.